Maximum midexpiratory flow (MMEF, FEF25-75%)

MMEF, FEF25-75%

The MMEF (maximum midexpiratory flow) is the average expiratory flow over the middle half of the FVC. It should be taken from the blow with the largest sum of FEV1 and FVC (procedures, ref. 2). This index is very highly correlated with FEF50%FVC, so that one index offers no advantage over the other. The MMEF is in use as an index of airway obstruction, but this is fraught with hidden dangers:

1. As MMEF = ½FVC/Δt (where Δt = time required to expire the middle half of the FVC), every disorder that affects the FVC will also influence the MMEF, just as in the case of MEFx%. During a forced expiration exhaled volume does not increase linearly with time; in the case of airway obstruction and reduced FVC the change in Δt, and therefore the severity of airway obstruction, is underestimated. It follows that the index is highly dependent on the validity of the FVC measurement and the level of expiratory effort.
2. The reverse situation arises after bronchodilatation: the improvement in airway patency is underestimated by an increase in the FVC. It is even possible that the MMEF remains the same or even decreases when the FVC and the FEV1 increase appreciably after a bronchodilator drug. Assessing the MMEF isovolumetrically (at the same lung volume), i.e. by timing the interval taken to exhale the middle half of the largest FVC recorded before and after the intervention can circumvent this problem. Commonly available instruments do not provide facilities to this end.

These pitfalls also apply to longitudinal observations, unless the FVC would remain the same. An improvement in MMEF after administration of a bronchodilator, if not accompanied by an improvement in FEV1 and/or VC, should be disregarded (ref. 1). If MMEF is measured serially this should be done isovolumetrically (ref. 2), otherwise e.g. an improvement in FEV1 can be associated with paradoxical falls in MMEF. Another problem is that, depending on age, sex and ethnic group, the between subject coefficient of variation varies between 20-62% for FEF25-75% (ref.3), severely limiting its usefulness for clinical purposes.

Measurements of these flows are not clinically useful
Many people hold the view that FEF25-75% and FEF75% are more sensitive to airflow limitation than FEV1 or FEV1/FVC, because unlike these flows the latter indices are not sufficiently sensitive to "small airways disease". This idea arose because McFadden and Linden, who first launched this idea, based their conclusions on the erroneous, yet still popular assumption, that 80% of predicted defines the lower limit of normal (LLN). Yet, in childhood the LLN is at 65% predicted and in those >80 years at 37%.
Consider the following. Both FEV1 and FEF25-75% are average flows over the first 1 second and the middle half of the FVC, respectively. For practical purposes they virtually share the same expiratory flows in healthy people and subjects with airways obstruction. It is difficult to conceive how this would affect the FEF25-75% and not the FEV1. One might argue that as the FEV1/FVC ratio falls progressively below 0.75 so that flows are shared over a smaller portion of the FVC manoeuvre, FEF25-75% might be more affected. However, FEV1/FVC and FEF25-75% are very highly correlated (ref. 4). Not surprisingly, a large study showed that neither FEF25-75% nor FEF75% added information that could not be derived from FEV1, FVC and FEV1/FVC. These findings were subsequently corroborated in additonal studies (ref. 4).